I Tested DeepSeek Agent Harness Against Codex and Claude Code

You are helping me turn one specific YouTube video into real, durable learning.

Source video:
- Title: I Tested DeepSeek Agent Harness Against Codex and Claude Code
- URL: https://www.youtube.com/watch?v=kFvEReALJL8
- Topic: Interfaces + Open Design
- My current learning frame: Install the DeepSeek harness, build one small custom UI plugin for a workflow you personally struggle to track, then run a multi-bug coding task through the coordinator-worker pattern to see parallel isolated sessions in action.
- Why this matters: New playlist item from Owain Lewis; queued for transcript-backed review, topic mapping, and a practical learning artifact.

Transcript anchors from this exact video:
- 0:44 / Evidence 1: "harness. So, when you click on this, you see a custom canband board that shows all of the work that I currently have running inside this agent harness. You can see here, this is the task I'm currently..."
- 2:15 / Evidence 2: "build basically anything on top of the agent harness. Okay. Okay, so if you do want to install this harness, it's pretty easy to set up. The first thing you want to do is just copy this one..."
- 4:25 / Evidence 3: "coding, you can do something like this. We can say, can you ping claude code and codeex? They're running on my local machine. and we're just going to basically call them via a tool call which allows you..."
- 6:45 / Evidence 4: "etc. But you can dispatch these tasks to individual agents. So for example, you can either work on this yourself or queue it up for an agent to work on. And you can store knowledge bases as well."
- 8:37 / Evidence 5: "agent to spawn these sessions in the sidebar as new sessions, the agent didn't think it had the tools available. I had to argue with it quite a lot and then pointed out that it does have RPC..."
- 10:12 / Evidence 6: "Very very similar. You can basically just see everything that the agent is doing. You can search through the you know the audit history. You just get more visibility into what your agents are doing. I'm not sure..."
- 13:13 / Evidence 7: "Check out the links in the description if you want to go deeper or if you want to spend more time going into agentic coding, building AI systems, or improving or growing your career or your business with..."

Video-aware target:
- Prompt lane: Hermes operations
- Mechanism to extract: Identify the operations control that makes long-running agent work visible, recoverable, or safer.
- Artifact to produce: A Hermes-style agent-ops runbook with health checks, state transitions, logs, recovery steps, and review criteria.
- Artifact must include: health check; state model; permission boundary; log source; recovery action

Your task:
1. Use the transcript anchors above as the primary source packet. If you add outside context, label it clearly as outside context and keep it secondary.
2. Create a source-check table with columns: timestamp, claim, transcript support, what the demo proves, confidence, and what still needs verification.
3. Extract the actual teachable mechanism from the video: Identify the operations control that makes long-running agent work visible, recoverable, or safer. Do not invent claims that are not supported by the title, lesson frame, or transcript anchors.
4. Build a reusable learning artifact: A Hermes-style agent-ops runbook with health checks, state transitions, logs, recovery steps, and review criteria.
5. Include:
   - a plain-English definition of the core idea
   - a diagram or structured model using this sequence: Project state -> Session -> Queue/Kanban -> Tools -> Logs -> Recovery -> Post-run review
   - answers to these source questions: What operational failure is prevented? | What state is visible? | What can be recovered or redirected?
   - 3 concrete examples that apply the video idea to real agentic work, such as Hermes Kanban triage; local model endpoint check; agent swarm recovery review
   - 2 failure modes the video helps prevent, chosen from the transcript evidence and these likely risks: treating UI features as reliability; missing logs; no stop/recover path
   - a checklist for the next real workflow, focused on: status, model/backend, tools, logs, recovery
   - one practical exercise with a clear done signal: Write a runbook for restarting one stuck Hermes-style agent session.
6. Add a "learning transfer" section: what changes in my workflow tomorrow if I actually learned this?
7. Add a "source check" section that cites which transcript anchor supports each major takeaway.

Quality bar:
- Make this specific to "I Tested DeepSeek Agent Harness Against Codex and Claude Code", not a generic Interfaces + Open Design essay.
- Ground each ops recommendation in transcript evidence about state, queues, models, tools, security, logs, or recovery.
- Prefer operational examples, failure modes, and reusable artifacts over broad definitions.
- Call out uncertainty instead of smoothing over weak evidence.
- Avoid these generic drifts: feature cheerleading; ops advice without logs/state; assuming reliability from a demo alone.
- If evidence is weak or missing, stop and say what transcript segment or timestamp needs review instead of guessing.
- Finish with a concise artifact I could paste into my learning app.